Emulson
Autonomous compounding and emulsification control.
Cosmetics & Personal Care
Serumon
Autonomous operations layer for cosmetics and personal-care manufacturing — autonomous compounding, shade vision, fill-weight control, batch assurance.
Overview
What Serumon does.
Serumon closes the loop on cosmetics and personal-care manufacturing: autonomous compounding, shade and quality vision, fill and weight control, stability and micro intelligence and the make-and-fill digital twin.
Every batch produces labelled compound, shade and fill traces that train the next model, tightening the loop from base to filled unit.
| Detail | Value |
|---|---|
| Stage | Seed |
| Category | Physical AI |
| Year | 2026 |
| Website | serumon.com |
| Headquarters | Not disclosed |
| Raised | $1M |
| Investor | 10xC |
| Model | Per-line software subscription with a batch-release outcome component. |
| Measured by | Batch-release time, shade conformance, fill-weight accuracy, lines expanded per customer. |
Products
One loop, module by module.
Every deployment on serumon.com writes labelled traces back into the model, so the system generalises across plants, alloys, grades and geometries.
Shadera
Shade and quality vision across batch and line.
Fillix
Fill and weight control.
Stabion
Micro and stability intelligence.
Twynex
Make-and-fill digital twin.
Why we invested
The thesis behind Serumon.
Cosmetics runs on brand shade and consistency: a batch that drifts a half-shade is a write-off, and compounding still leans on the person who knows the batch. Personal-care volume is expanding faster than the people who run it.
Serumon holds the shade in software. The buyer is a cosmetics or personal-care manufacturer paying in hard currency; the wedge is one compounding or fill step, and batch-release time is the number that opens the next module.
Who is building it
Saad Khan
Co-Founder & CEOSaad owns the brand relationships, the pilot programme and the commercial model.
Alizey Shah
Co-Founder & CTOAlizey owns the compounding and vision models, the edge runtime and the batch-assurance architecture.
Building something similar?
Building in Physical AI? We want to hear about it.